[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Translating required-cleanup and draft-comment
Hi, The translation subcommittee was asked when required-cleanup and draft-comment should be translated. The confusion comes from two conflicting statements in the spec. First, we state that these elements should not be translated by default, because they will not appear in the output. Second, we state that the translate attribute is always inherited, and should take priority over defaults for an element. The following case is then unclear: <p translate="yes">Translatable text <required-cleanup>fix this later</required-cleanup> </p> The required-cleanup tag inherits translate="yes" even though the content should not be translated. Essentially, we state that required-cleanup should have a default of translate="no", but because the default is not explicit, tools do not work automatically. To deal with these issues, the translation subcommittee has made the following recommendations: 1. Add a default of translate="no" to required-cleanup and draft-comment in the DTDs and Schemas. 2. Clarify the spec to indicate that the translate attribute is respected on and within these elements. If translate is set to "yes" on required-cleanup, the contents should be translated. If translate is set to "yes" on a paragraph within required-cleanup, the contents of that paragraph should be translated. In other words, the standard definition of the translate attribute still applies. 3. Update the documentation for the 2 elements to indicate that the contents will not be translated unless the translate attribute is set to "yes". This recommendation recognizes that there are cases where the content of these elements may be translated. For example, the element may be specialized into a reusableObject element, where the content must be translated before it is pulled into other topics. The documentation should clarify that this can be handled by giving the specialized element a default of translate="yes" inside the document type. Sound reasonable? Are there any problems with this change? Thanks, Robert D Anderson IBM Authoring Tools Development Chief Architect, DITA Open Toolkit (507) 253-8787, T/L 553-8787
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]